INTERNET-DRAFT Larry Masinter
Adobe Systems Incorporated
Martin Duerst
draft-masinter-url-i18n-07.txt W3C/Keio University
Expires July 2001 January 4, 2001
Internationalized Resource Identifiers (IRI)
Status of this Memo
This document is an Internet-Draft and is in full conformance with all
provisions of Section 10 of RFC2026.
Internet-Drafts are working documents of the Internet Engineering Task
Force (IETF), its areas, and its working groups. Note that other
groups may also distribute working documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet- Drafts as reference
material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
This document is not a product of any working group, but may be
discussed on the mailing list <uri@w3.org>. For more information on
the topic of this internet-draft, please also see [W3C IRI].
Abstract
This document defines a new protocol element, an Internationalized
Resource Identifier (IRI). An IRI is a sequence of characters from
the Universal Character Set [10646]. A mapping from IRIs to URIs
[RFC 2396] is defined, which means that IRIs can be used instead
of URIs where appropriate to identify resources.
Defining a new protocol element was preferred to extending or
changing the definition of URIs to allow a clear distinction and to
avoid incompatibilities with existing software. Guidelines for the
use and deployment of IRIs in various software components that now
deal with URIs are provided.
1. Introduction
1.1 Overview and Motivation
A URI is defined in [RFC2326] as a sequence of characters chosen from
a limited subset of the repertoire of ASCII characters. This document
defines a new protocol element, an IRI (Internationalized Resource
Identifier) by extending the syntax of URIs to a much wider repertoire
of characters. It also defines "internationalized" versions corresponding
to other constructs from [RFC2326].
The characters in URIs are frequently used for representing words of
natural languages. Using words from natural languages in URIs is very
common. Such usage has many advantages: such URIs are easier to
memorize, easier to interpret, easier to transcribe, easier to create,
and easier to guess. For most languages other than English, however,
the natural script uses characters other than A-Z. For many people,
handling Latin characters is as difficult as handling the characters
of other scripts is for people who use only the Latin alphabet. Many
languages with non-Latin scripts do have transcriptions to Latin
letters and such transcriptions are now often used in URIs, but they
introduce additional ambiguities.
The infrastructure for the appropriate handling of characters from
local scripts is now widely deployed in local versions of operating
system and application software; and software that can handle a wide
variety of scripts and languages at the same time is increasing in
use.
Using characters outside of A-Z in IRIs brings with it some
difficulties; a discussion of drawbacks and necessary workarounds can
be found in the later sections of this document.
URIs often contain Internet host names embedded with them. There is an
ongoing discussion of internationalization and host names; the
specific issues of the relationship of IRIs and possible future
"internationalized" host names are not discussed here.
The definition of IRIs here is consistent with the URN syntax
[RFC2141] as well as recent URL scheme definitions that define
encodings of non-ASCII characters based on (e.g., IMAP URLs [RFC 2192]
and POP URLs [RFC 2384]). In these examples, the representation of
non-ASCII characters is based on the same mechanism (using UTF-8 and
hex encoding, as recommended in [RFC 2718]) as the conversion from
IRIs to URIs defined below. As a consequence, when software components
are upgraded to handle IRIs, the non-ASCII characters escaped e.g. in
in a POP URIs will appear directly as characters in the corresponding
IRI.
1.2 Definitions
The following definitions are used in this document; they follow the
terms in [RFC 2130] and [RFC2277]:
character An abstract object with a separate identity.
"LATIN CAPITAL LETTER A" is a character.
octet 8 bits
character repertoire A set of characters (in the Mathematical
sense)
sequence of characters A sequence (one after another) of characters
sequence of octets A sequence (one after another) of octets
character encoding A method of representing a sequence of
characters as a sequence of octets (maybe
with variants). A method of (unambiguously!)
converting a sequence of octets into a
sequence of characters.
code point A placeholder for a character in a character
encoding, for example to encode additional
characters in future versions of the
character encoding.
charset The name of a parameter or attribute used
to identify a character encoding.
2. IRI Syntax
This section defines the syntax of Internationalized Resource
Identifiers (IRIs).
As with URIs, an IRI is defined as a sequence of characters. This
definition accommodates the fact that IRIs may be written on paper or
read over the radio as well as being transmitted over the network.
Also, the same IRI may be represented as different sequences of octets
in different protocols or documents if these protocols or documents
use different character encodings. Using the same character encoding
as the containing protocol or document assures that the characters in
the IRI can be handled (among else displayed) in the same way as the
rest of the protocol or document.
2.1 Summary of IRI syntax
IRIs are defined similarly to URIs in [RFC2386] (as modified by
[RFC2732]), but the class of unreserved characters is extended by
adding all the characters of the UCS (Universal Character Set,
[ISO10646]) beyond U+0080, subject to the limitations given in
Section 3.
Otherwise, the syntax and use of components and reserved characters is
the same as that in [RFC2396]. All the operations defined in
[RFC2396], such as the resolution of relative URIs, can be applied to
IRIs by IRI-processing software exactly in the same way as this is
done by URI-processing software.
Characters outside the ASCII range MUST NOT be used for syntactical
purposes such as to delimit components in newly defined schemes. As an
example, it is not allowed to use U+00A2, CENT SIGN, as a delimiter,
because it is in the iunreserved category, in the same way as it is
not possible to use '-' as a delimiter, because it is in the
unreserved category.
2.2 ABNF for IRIs
While it might be possible to define IRIs merely by their
transformation to URIs, it is possible that IRIs are accepted and
processed directly in some situations. For this reason, an ABNF
definition for IRIs is given here.
IRI-reference = [ absoluteIRI | relativeIRI ] [ "#" fragment ]
absoluteIRI = scheme ":" ( hier_part | opaque_part )
relativeIRI = ( inet_path | iabs_path | irel_path )
[ "?" iquery ]
ihier_part = ( inet_path | iabs_path ) [ "?" i-query ]
iopaque_part = iric_no_slash *iric
iric_no_slash = iunreserved | escaped | ";" | "?" | ":" | "@" |
"&" | "=" | "+" | "$" | ","
inet_path = "//" iauthority [ iabs_path ]
iabs_path = "/" ipath_segments
irel_path = irel_segment [ iabs_path ]
irel_segment = 1*( iunreserved | escaped |
";" | "@" | "&" | "=" | "+" | "$" | "," )
iauthority = server | ireg_name
ireg_name = 1*( iunreserved | escaped | "$" | "," |
";" | ":" | "@" | "&" | "=" | "+" )
ipath_segments = isegment *( "/" isegment )
isegment = *ipchar *( ";" iparam )
iparam = *ipchar
ipchar = iunreserved | escaped |
":" | "@" | "&" | "=" | "+" | "$" | ","
iquery = *iric
ifragment = *iric
iric = reserved | iunreserved | escaped
iunreserved = ichar | unreserved
ichar = << any character of UCS [ISO10646] beyond
U+0080, subject to limitations in Section
3.1. >>
The following are the same as [RFC2396] as modified by [RFC2732]:
reserved = ";" | "/" | "?" | ":" | "@" | "&" | "=" | "+" |
"$" | "," | "[" | "]"
unreserved = alphanum | mark
mark = "-" | "_" | "." | "!" | "~" | "*" | "'" |
"(" | ")"
escaped = "%" HEXDIG HEXDIG
server = [ [ userinfo "@" ] hostport ]
userinfo = *( unreserved | escaped |
";" | ":" | "&" | "=" | "+" | "$" | "," )
hostport = host [ ":" port ]
host = hostname | IPv4address | IPv6reference
ipv6reference = "[" IPv6address "]"
hostname = *( domainlabel "." ) toplabel [ "." ]
domainlabel = alphanum | alphanum *( alphanum | "-" ) alphanum
toplabel = alpha | alpha *( alphanum | "-" ) alphanum
IPv6address = hexpart [ ":" IPv4address ]
IPv4address = 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT
IPv6prefix = hexpart "/" 1*2DIGIT
hexpart = hexseq | hexseq "::" [ hexseq ] | "::"
[ hexseq ]
hexseq = hex4 *( ":" hex4)
hex4 = 1*4HEXDIG
port = *DIGIT
scheme = alpha *( alpha | digit | "+" | "-" | "." )
alphanum = alpha | digit
alpha = lowalpha | upalpha
lowalpha = "a" | "b" | "c" | "d" | "e" | "f" | "g" | "h" | "i" |
"j" | "k" | "l" | "m" | "n" | "o" | "p" | "q" | "r" |
"s" | "t" | "u" | "v" | "w" | "x" | "y" | "z"
upalpha = "A" | "B" | "C" | "D" | "E" | "F" | "G" | "H" | "I" |
"J" | "K" | "L" | "M" | "N" | "O" | "P" | "Q" | "R" |
"S" | "T" | "U" | "V" | "W" | "X" | "Y" | "Z"
2.3 Mapping of IRIs to URIs
This section defines how to convert an IRI to a URI. This mapping has
two purposes:
a) Syntax limitation: IRIs are valid only if they map to
syntactically valid URIs.
b) Interpretation: The "meaning" of a IRI--the way than an IRI can be
used to identify a resource -- is defined by the mapping: the IRI
is translated to a URI which is then used to identify (name,
locate, etc.) the appropriate resource.
This mapping can be accomplished by the following steps:
1) Represent the IRI characters as a sequence of characters from the
UCS.
2) Normalize the character sequence according to Normalization Form
C, as defined in [IETFNorm]. (See further discussion in Section
3.1.)
3) For each character that is not in the US-ASCII repertoire, apply
the following:
3.1) Convert the character to a sequence of one or more octets
using UTF-8 [RFC 2279]. Note: All resulting octets will have
values greater than 127.
3.2) Convert each octet to %HH, where HH is the hexadecimal
notation of the octet value. Note: This is identical to the
escaping mechanism in Section 2.4.1 of [RFC 2396].
3.3) Replace the original character by the resulting character
sequence.
In step 3), octets allowed in URIs MUST NOT be escaped further,
because they are already in their correct escaping stage in IRIs.
The above mapping produces an URI fully conforming to [RFC 2396] out
of each IRI. In addition, it results in the identity transformation
for URIs. Every URI is therefore by definition an IRI. (This is due to
the fact that Normalization Form C does not modify strings of US-ASCII
characters.)
2.4 Converting URIs to IRIs
In some situations, it might be desirable to convert a URI into an IRI
which would be equivalent, especially if it is known that the URI
originally resulted from transformation from an IRI, or the URI is
an URN or an URL using UTF-8 to encode non-ASCII characters (see
last paragraph of Subsection 1.2).
However, it is important to note that in the general case, this
conversion is not reliable. Many URI schemes and URI components allow
arbitrary octets to be encoded. Even when the conversion is successful
because the hex-escaped octets in the URI are legal in UTF-8, this may
be so just by accident. Due to the regularities in the octet patterns
of UTF-8, it is likely that the result is appropriate, but there
cannot be a guarantee that the IRI that results is sensible or
corresponds to any original meaning.
Conversion from an URI to an IRI is done using the following steps:
1) Convert all hexadecimal escapes (% followed by two hexadecimal
digits) of %80 and higher to the corresponding octets. (The
result is a sequence of intermixed 'characters' and 'octets'; it
is necessary to distinguish strictly between characters and
octets in this procedure.)
2) Convert all sequences of octets that are strictly legal UTF-8
sequences to the corresponding sequences of characters. Note: The
properties of UTF-8 make sure that this will in all cases lead to
the same result.
3) Convert back to octets all resulting characters that either do
not meet the syntax limitations of Section 3.1 or are identified
as unadvisable.
4) Convert all the remaining octets (i.e. those not affected by step
2 and those produced by step 3) back to hexadecimal escapes.
This procedure will convert as many escaped non-ASCII characters as
possible to characters in an IRI. In some cases, the results will be
that only some components of the URI will become unescaped in the
resulting IRI, while others will remain escaped.
3. Considerations for use of IRIs
3.1 IRI Character Limitations
Not all characters of the UCS are appropriate for use as resource
identifiers. This section gives the limitations on characters and
character sequences usable for IRIs.
a. The repertoire of characters allowed in each IRI component is
limited by the definition of that component. For example, the
definition of host names in URIs does not allow hex escapes, or
"_", or many other punctuation characters. This specification
does not relax those limits, and so IRIs may not contain any
non-ASCII characters in host names. The scheme component is
likewise not extended beyond ASCII.
b. Many of the limitations of the repertoire of URIs were selected
because of the requirements for transcription: URIs should not
contain characters that are difficult for a person to transcribe
from a visible representation of the character into typed text;
characters that are likely to be used to delimit URIs in text
and print were likewise avoided.
For this reason, UCS characters similar to those in the categories
"space", "delims", and "unwise" in [RFC 2396, Section 2.4.3] should
not be used. [More details to be added.]
c. The UCS contains many areas of "characters" which have no simple
way of typing them. These should be avoided. Characters that
fall into this category include Dingbats, Mathematical
and other symbols, ligatures and presentation forms.
d. The UCS contains many areas of characters for which there are
strong visual look-alikes. Because of the likelihood of
transcription errors, these also should be avoided. This includes
the full-width equivalents of ASCII characters, and half-width
Katakana characters for Japanese, and many others. Please note
that the likelihood of transcription errors may depend on the
context.
e. Characters with no visual representation may not be interoperably
entered. "Control characters" MUST NOT be used. This includes
the traditional ranges of control characters (U+0000-U+001F and
U+0080-U+009F) as well as swapping and plane-14 language tag
characters. [More details to be added.]
f. Some code points are reserved for private use or for special
encoding purposes. They are not interoperable. Code points reserved
for private use MUST NOT be used. Code points reserved for
surrogates MUST NOT be used.
g. Where there exist duplicate ways of encoding a certain character as
visible to the user, Normalization Form C as defined in [IETFNorm]
MUST be used.
For reasons of transcribability, many characters have been
excluded above. These can nevertheless be encoded in an IRI if
necessary. However, they have to be escaped using the procedure
in Section 2.3. For example, a space has to be encoded in an
URI and in an IRI as %20. A non-breaking space (U+00A0) can be
encoded as %C2%A0.
3.2 Bidirectional IRIs for Right-to-Left languages
Bidirectional (BIDI) IRIs are IRIs containing characters with an
inherent right-to-left writing direction; these require additional
attention when being converted from a visual representation to a
digital representation and back.
In digital representations (as well as when read/spelled), the
sequence of components and characters should be in "logical" order:
This conforms to the specifications for the UCS and allows generic
operations, such as the resolution of relative IRIs, to be carried
out without special provisions.
A visual representation placing the IRI characters strictly from left
to right would make some of its components, such as words written in
Arabic or Hebrew, unreadable. On the other hand, an uncontrolled
reversion of the whole IRI would make components with Latin or other
left-to-right words unreadable, and/or would obscure the sequence of
the IRI components. In addition, a direct application of the Unicode
bidirectionality algorithm [UNIV3, Section 3.12] would relocate the
reserved characters that define the structure of an URI because most
of them have neutral directionality.
The visual representation of IRIs is therefore defined as follows:
- The IRI as a whole is presented from left to right, component by
component. Components of an IRI are parts of an IRI that are
delimited by reserved characters.
- Within each component, the Unicode bidi algorithm is applied,
assuming a left-to-right embedding context.
For display, this behavior can be achieved by preceding the IRI with
an LRE (left to right embedding) character, following it with a PDF
(pop directional formatting) character, and preceding and following
each reserved character by an LRM (left to right mark) character. In
this form, it can be passed to a display engine supporting the Unicode
BIDI algorithm.
We are requesting feedback on the this approach from the affected
communities.
3.3. Processing relative IRIs
Processing of relative forms of IRIs against a base is handled
straightforwardly; the algorithms of RFC 2396 may be applied directly.
Another possibility is to convert both the relative IRI and the base
IRI to URIs, then absolutize, and then convert back. If only one of
the involved identifiers is an IRI, it should also be converted to
an URI.
4. Software requirements
This section explains the issues and difficulties in supporting IRIs
in the same software components and operations that currently process
URIs: software interfaces that handle URIs, software that allows users
to enter URIs, software that generates URIs, software that displays
URIs, formats that transport URIs, and software that interprets URIs
may all require significant modification before functioning properly
with IRIs.
4.1 URI/IRI software interfaces
Software interfaces that handle URIs, such as URI-handling APIs and
protocols transferring URIs, may need modifications before they can
handle IRIs.
Note that although an IRI is defined as a sequence of characters,
software interfaces for URIs typically function on sequences of
octets. Thus, it is necessary to define clearly which character
encoding is used.
In case the current handling is based on ASCII, UTF-8 should be chosen
as the encoding for IRIs, because this is compatible with ASCII, is in
accordance with the recommendations of [RFC 2277], and makes it easy
to convert to URIs where necessary. In any case, the encoding used must not
be left undefined.
Intermediate software interfaces between IRI-capable components and
URI-only components must map the IRIs as per section 2.3 above, when
transfering from IRI-capable to URI-only components. The transfer
in the other direction requires no mapping, although the mapping
described in section 2.4 above may be performed.
It is preferable to not to perform this inverse mapping when there is
a chance that this cannot be done correctly. For example, if it cannot
be checked whether the sequence of %HH escapes corresponds to a valid
sequence of UTF-8 octets, unescaping should not be done.
4.2 URI/IRI entry
There are components that allow users to enter URIs into the system,
e.g., by typing or dictation. This software must be updated to allow
for IRI entry.
A person viewing a visual representation of an IRI (as a sequence of
glyphs, in some order, in some visual display) or hearing an IRI,
will use a keyboard entry method for keys in that language to input
the IRI. Depending on the script and the input method used, this may
be a more or less complicated process.
The process of IRI entry must assure, as far as possible, that the
limitations defined in Section 3.1 are met. This may be done by
choosing appropriate input methods or variants thereof, by
appropriately converting the characters being input, by eliminating
characters that cannot be converted, and/or by issuing a warning or
error message to the user.
An input field primarily or only used for the input of URIs/IRIs
should allow the user to view an IRI as converted to an URI. Places
where the input of IRIs is frequent should provide the possibility for
viewing an IRI as converted to an URI. This will help users when some
of the software they use does not yet accept IRIs.
An IRI input component that interfaces to components that handle URIs,
but not IRIs, must escape the IRI before passing it to such a
component.
The input of IRIs with right-to-left characters requires additional
care to keep the visual and the internal representation in synch, and
to eliminate control characters and marks used to control the display
before passing the IRI over to a resolver. IRI input fields that allow
the input of right-to-left characters MUST provide this functionality. IRI
input fields that do not provide this functionality MUST NOT allow
the input of right-to-left characters.
4.3 URI/IRI generation
Systems that are offering resources through the Internet, where those
resources have logical names, sometimes automatically generate URIs
for the resources they offer. For example, some HTTP servers can
generate a 'directory listing' for a file directory under their purview,
and then respond to the generated URIs with the files.
Many legacy character encodings are in use in various file systems.
Many currently deployed systems do not transform the local character
representation of the underlying system before generating URIs.
For maximum interoperability, systems that generate resource
identifiers should do the appropriate transformations. They should use
IRIs converted to URIs in cases where it cannot be expected that the
recipient is able to handle IRIs. Due to the way most user agents currently
work, native IRIs, encoded in UTF-8, may be used if the
recipient announces that it can interpret UTF-8.
This recommendation in particular applies to HTTP servers. For FTP
servers, similar considerations apply, see in particular [RFC 2640].
4.4 URI/IRI selection
In some cases, resource owners and publishers have control over the
IRIs used to identify their resources. Such control is mostly
executed by controlling the resource names, such as file names,
directly.
In such cases, it is recommended to avoid choosing IRIs that are
easily confused. For example, for ASCII, the lower-case ell "l" is
easily confused with the digit one "1", and the upper-case oh "O" is
easily confused with the digit zero "0". Publishers should avoid to
unintentionally confuse users with "br0ken" or "1ame" identifiers.
Outside of the ASCII range, there are many more opportunities for
confusion; a complete set of guidelines is too lengthy to include
here. As long as names are limited to characters from a single script,
native writers of a given script or language will know best when
ambiguities can appear, and how they can be avoided. What may look
ambiguous to a stranger may be completely obvious to the average
native user.
Note that the limitations defined in Section 3.1 and the
recommendations given here are of a different nature. The limitations
defined in Section 3.1 are necessary to avoid duplicate encodings that
are artifacts of digital representation and that the user has no way
to distinguish visually. On the other hand, in a given context, an
identifier such as "BOX0021" can be completely appropriate, and it is
impossible to find an algorithm that distinguishes the appropriate
from the confusing identifiers.
In certain cases, there is a chance that letters from different
scripts look the same. The best know example is the Latin 'A', the
Greek 'Alpha', and the Cyrillic 'A'. To disambiguate such cases, it
should be assumed that all letters in a component are from the same
script. This is similar to the heuristics used to distinguish between
letters and numbers in the examples above. Also, for the above three
scripts, using lower case letters results in much fewer ambiguities
than using upper-case letters.
4.5 Display of URI/IRIs
Many systems contain software that presents URIs to users as part of
the system's user interface (sometimes presenting 'friendly' URIs,
i.e., a shortened or more legible subset of the URI.) This section
applies to this presentation, as well as to the strategy for printing
URIs in magazines, newspapers, or reading them over the radio.
Software that displays identifiers to users should follow a general
principle: "Don't display something to a user that the user would not
be able to enter." The consequences of this principle require
judgement about the availability of software that implements the entry
methods described in Section 3.2.
a) In situations where a viewer is not likely to have software that
implements non-ASCII character entry (as described in Section 3.1),
or where it can be expected that only a limited range of non-ASCII
characters can be entered, any part of an IRI containing characters
outside the range allowed in [RFC 2396] or any additions should be
escaped before being displayed.
b) In situations where a viewer _is_ likely to have such software,
IRIs may be displayed directly.
For display of BIDI IRIs, please see section 3.2.
4.6 Interpretation of URI/IRIs
Software that interprets IRIs as the names of local resources should
accept IRIs in multiple forms, and convert and match them with the
appropriate local resource names.
First, multiple representations includes both IRIs in the native
character encoding of the protocol (UTF-8 if not otherwise defined)
and also their URI counterparts.
Second, it may include URIs constructed based on other character
encodings than UTF-8. Such URIs may be produced by user agents that do
not conform to this specification and use legacy encodings to convert
non-ASCII characters to URIs. Whether this is necessary, and what
character encodings to cover, depends on a number of factors, such as
the legacy character encodings used locally and the distribution of
various versions of user agents. For example, software for Japanese
may accept URIs in Shift_JIS and/or EUC-JP in addition to UTF-8.
Third, it may include additional mappings to be more user-friendly and
robust against transmission errors. These would be similar to how
currently some servers treat URIs as case-insensitive, or perform
additional matchings to account for spelling errors. For characters
beyond the ASCII repertoire, this may e.g. include ignoring the
accents on received IRIs or resource names where appropriate. Please
note that such mappings, including case mappings, are
language-dependent.
It may seem to be difficult to unambiguously identify a resource if
too many mappings are taken into consideration. This can indeed be the
case. However, IRIs escaped to URIs and non-escaped IRIs can always
clearly be distinguished. Also, the regularity of UTF-8 makes the
potential for collisions lower than it may seem at first sight.
4.7 Transportation of URI/IRIs in document formats
Document formats that transport URIs may need to be upgraded to allow
the transport of IRIs. In those cases where the document as a whole
has a native character encoding, IRIs should also be encoded in this
encoding, and converted accordingly by a parser or interpreter. IRI
characters that are not expressible in the native encoding should be
escaped according to Section 2.2, or may be escaped in another way if
the document format provides a way to do this. For example, in HTML,
XML, or SGML, numeric character references can be used.
Please note that an interpretation of characters in URIs outside the
ASCII repertoire as IRIs, conforming to this specification, is already
defined as error behavior in HTML 4.0 [HTML4] and in XML 1.0 [XML1].
Also, it is under discussion to require this behavior from all W3C
formats [CharMod].
5. Upgrading strategy
As this recommendation places further constraints on software for
which many instances are already deployed, it is important to
introduce upgrades carefully, and to be aware of the various
interdependencies.
If IRIs cannot be interpreted correctly, they should not be generated
or transported. This suggests that upgrading URI interpreting software
to accept IRIs should have highest priority.
On the other hand, a single IRI is interpreted only by a single or
very few interpreters that are known in advance, while it may be
entered and transported very widely.
Therefore, IRIs benefit most from a broad upgrade of software to be
able to enter and transport IRIs, but before publishing any
individual IRI, care should be taken to upgrade the corresponding
interpreting software in order to cover the forms expected to be
received by various versions of entry and transport software.
The upgrade of generating software to generate IRIs instead of a local
encoding should happen only after the service is upgraded to accept
IRIs. Similarly, IRIs should only be generated when the service
accepts IRIs and the intervening infrastructure and protocol is known
to transport them safely.
Display software should be upgraded only after upgraded entry software
has been widely deployed to the population that will see the displayed
result.
These recommendations, when taken together, will allow for the
extension from URIs to IRIs in order to handle scripts other than
ASCII while minimizing interoperability problems.
6. Security Considerations
If IRI entry software normalizes the characters entered, but the
resource names on the interpreting side are not normalized
accordingly, and the interpreting software does not take this into
account, there is a possibility of "spoofing". Similar possibilities
turn up when interpreting software accepts URIs in various native
encodings or allows accents and similar things to be ignored.
"Spoofing" means that somebody may add a resource name that looks the
same or similar to the user while actually being different, or a
resource name that contains the same characters, but in a different
encoding. The added resource may pretend to be the real resource by
looking very similar, but may contain all kinds of changes that may be
difficult to spot but can cause all kinds of problems.
Conceptually, this is no different from the problems surrounding the
use of case-insensitive web servers. For example, a popular web page
with a mixed case name (http://big.site/PopularPage.html) might be
"spoofed" by someone who obtains access to
(http://big.site/popularpage.html).
However, the introduction of character normalization, of additional
mappings for user convenience, and of mappings for various encodings
may increase the number of spoofing possibilities. In some cases, in
particular for Latin-based resource names, this is usually easy to
detect because UTF-8-encoded names, when interpreted and viewed as
legacy encodings, produce mostly garbage. In other cases, when
concurrently used encodings have a similar structure, but there are no
characters that have exactly the same encoding, detection is more
difficult. A good example may be the concurrent use of Shift_JIS and
EUC-JP on a Japanese server.
Administrators of large sites which allow independent users to
create subareas may need to be careful that the aliasing rules
do not create chances for spoofing.
7. Acknowledgements
The issue addressed here has been discussed at numerous times over the
last many years; for example, there was a thread in the HTML working
group in August 1995 (under the topic of "Globalizing URIs") in the
www-international mailing list in July 1996 (under the topic of
"Internationalization and URLs"), and ad-hoc meetings at the Unicode
conferences in September 1995 and September 1997.
Thanks to Francois Yergeau, Chris Wendt, Yaron Goland, Graham Klyne,
Roy Fielding, M.T. Carrasco Benitez, James Clark, Andrea Vine, and
many others for help with understanding the issues and possible
solutions. Thanks also to the members of the W3C I18N Working Group
and Interest Group for their work on [CharMod]. Special thanks to
Makoto MURATA for regularly reminding us when an update was due.
8. Copyright
Copyright (C) The Internet Society, 1997. All Rights Reserved.
This document and translations of it may be copied and furnished to
others, and derivative works that comment on or otherwise explain it
or assist in its implementation may be prepared, copied, published
and distributed, in whole or in part, without restriction of any
kind, provided that the above copyright notice and this paragraph
are included on all such copies and derivative works. However, this
document itself may not be modified in any way, such as by removing
the copyright notice or references to the Internet Society or other
Internet organizations, except as needed for the purpose of
developing Internet standards in which case the procedures for
copyrights defined in the Internet Standards process must be
followed, or as required to translate it into languages other
than English.
The limited permissions granted above are perpetual and will not be
revoked by the Internet Society or its successors or assigns.
This document and the information contained herein is provided on an
"AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING
TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE."
9. Author's addresses
Larry Masinter
Adobe Systems Incorporated
Mail Stop W14
345 Park Ave
San Jose, CA 95110
mailto: LMM@acm.org
http://larry.masinter.net
Tel: +1 408 536-3024
Martin J. Duerst
W3C/Keio University
5322 Endo, Fujisawa
252-8520 Japan
duerst@w3.org
http://www.w3.org/People/D%C3%BCrst/
Tel/Fax: +81 466 49 1170
Note: The homepage URI of the second author is the escaped
form of an IRI.
Note: Please write "Duerst" with u-umlaut wherever
possible, e.g. as "Dürst" in XML and HTML.
10. References
[CharMod] M. Duerst and F. Yergeau, Ed., Character Model for the World
Wide Web, <http://www.w3.org/TR/charmod>, work in progress.
[HTML4] "HTML 4.0", World Wide Web Consortium,
<http://www.w3.org/TR/REC-html40/appendix/notes.html#h-B.2>.
[ISO10646] ISO/IEC, Information Technology - Universal Multiple-Octet
Coded Character Set (UCS) - Part 1: Architecture and Basic
Multilingual Plane, Oct. 2000, with amendments.
[IETFNorm] M. Duerst, M. Davis, "Character Normalization in IETF
Protocols", Internet Draft, September 2000,
<http://www.ietf.org/internet-drafts/draft-duerst-i18n-norm-04.txt>,
work in progress.
[RFC 2119] S. Bradner, "Key words for use in RFCs to Indicate
Requirement Levels", March 1997.
[RFC 2141] R. Moats, "URN Syntax", May 1997.
[RFC 2192] C. Newman, "IMAP URL Scheme", September 1997.
[RFC 2277] H. Alvestrad, "IETF Policy on Character Sets and
Languages".
[RFC 2279] F. Yergeau. "UTF-8, a transformation format of ISO 10646.",
January 1998.
[RFC 2384] R. Gellens, "POP URL Scheme", August 1998.
[RFC 2396] T.Berners-Lee, R.Fielding, L.Masinter. "Uniform Resource
Identifiers (URI): Generic Syntax." August, 1998.
[RFC 2616] R.Fielding, J.Gettys, et al, "Hypertext Transfer Protocol
-- HTTP/1.1", June 1999.
[RFC 2640] B. Curtis, "Internationalization of the File Transfer
Protocol", July 1999.
[RFC 2718] L. Masinter, H. Alvestrand, D. Zigmond, R. Petke,
"Guidelines for new URL Schemes", November 1999.
[RFC 2732] R. Hinden, B. Carpenter, L. Masinter, "Format for Literal
IPv6 Addresses in URL's", December 1999.
[UNIV3] The Unicode Consortium, "The Unicode Standard Version 3.0",
Addison-Wesley, Reading, MA, 2000.
[UNI15] M.Davis and M.Duerst, "Unicode Normalization Forms", Unicode
Technical Report #15, November 1999.
<http://www.unicode.org/unicode/reports/tr15/>
[W3C IRI] Internationalization - URIs and other identifiers
<http://www.w3.org/International/O-URL-and-ident.html>.
[XML1] "XML 1.0", World Wide Web Consortium Recommendation,
<http://www.w3.org/TR/REC-xml#sec-external-ent>.